INDEX

Explanations

phrases that indicate intentions or plans

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

thâu

-0.74

 <>",

-0.73

principalTable

-0.73

 esternos

-0.69

InjectAttribute

-0.68

 '\\;'

-0.68

StoryboardSegue

-0.66

 فريبيس

-0.66

 nakalista

-0.65

AddHtmlAttribute

-0.64

POSITIVE LOGITS

 quote

0.85

 paraphrase

0.77

me

0.68

sum

0.68

To

0.64

To

0.63

 illustrate

0.62

 summarize

0.60

 recap

0.59

 reiterate

0.56

Activations Density 0.123%

phrases that indicate intentions or plans

No Comments

No Known Activations